Dr. Anton Van den Hengel

Queen Mary Vision Laboratory seminar

Dr. Anton Van den Hengel
Department of Computer Science
University of Adelaide (Australia)

Date: Wednesday 24th September
Time: 2-3pm
Venue: Room CS 338

“Extending approximated maximum likelihood estimators for Computer Vision”

Parameter estimation on the basis of image-based measurements is a
central problem in computer vision.It arises in the estimation of
entities such as the fundamental matrix, homographies, and the trifocal
tensor amongst others.There have been a number of competing methods
developed for particular estimation problems and with different
statistical underpinnings.A subset of these schemes, including FNS,
CFNS, HEIV and renormalisation, are based on the minimisation of a cost
function which is of a particular form.This talk outlines a method by
which this class of methods may be extended to cover more general cost
functions. One result of this extension is that it allows the entire
class of methods to be placed within a common framework.